System and method for patient health data prediction using knowledge graph analysis

Abstract

A system and method for patient health data prediction and analysis which utilizes an automated text mining tool to automatically format ingested electronic health record data to be added to a knowledge graph, which enriches the edges between nodes of the knowledge graph with fully interactive edge data, which can extract a subgraph of interest from the knowledge graph, and which analyzes the subgraph of interest to generate a set of variables that define the subgraph of interest. The system utilizes a knowledge graph and data analysis engine capabilities of the data platform to extract deeper insights based upon the enriched edge data.

What is claimed is: 
     
       1. A system for patient health data prediction and analysis, comprising:
 a computing device comprising a memory and a processor; 
 a database comprising biochemical entities; 
 an automated text mining tool comprising a first plurality of programming instructions stored in the memory and operating on the processor of, wherein the first plurality of programming instructions, when operating on the processor, causes the computing device to:
 receive an electronic health record; 
 scrape the electronic health record to identify one or more biochemical entities that match one or more biochemical entities in the database; and 
 logically link the one or more identified biochemical entities to the one or more matched biochemical entities; 
 parse the information contained in the electronic health record into a standard data format; and 
 add the information contained in the standard data format to a biochemical knowledge graph; and 
 
 a data analysis engine comprising a first plurality of programming instructions stored in the memory and operating on the processor of, wherein the first plurality of programming instructions, when operating on the processor, causes the computing device to:
 receive a subgraph query, the subgraph query comprising at least one subgraph parameter; 
 retrieve from the biochemical knowledge graph a subgraph comprising all information related to the at least one subgraph parameter; 
 perform cluster analysis on the subgraph to identify one or more partitions within the subgraph; and 
 use a plurality of machine and deep learned classification models on the one or more partitions to identify a set of hyperparameters that define the partition. 
 
 
     
     
       2. A method for patient health data prediction and analysis, comprising the steps of:
 receiving an electronic health record; 
 scraping the electronic health record to identify one or more biochemical entities that match one or more biochemical entities in the database; 
 logically linking the one or more identified biochemical entities to the one or more matched biochemical entities; 
 parsing the information contained in the electronic health record into a standard data format; 
 adding the information contained in the standard data format to a biochemical knowledge graph; 
 receiving a subgraph query, the subgraph query comprising at least one subgraph parameter; 
 retrieving from the biochemical knowledge graph a subgraph comprising all information related to the at least one subgraph parameter; 
 performing cluster analysis on the subgraph to identify one or more partitions within the subgraph; and 
 using a plurality of machine and deep learned classification models on the one or more partitions to identify a set of hyperparameters that define the partition.